COHESION AND SUPPRESSION. 257 



affording a transition to the genus Gelsia whicb. only 

 •differs from Verbascum in having four didynamous 

 stamens, instead of five more or less equal ones. Camus 

 found suppression of one of the two stamens in various 

 species of Veronica. 



Warming and Battandier both found in the corn- 

 poppy {Papaver Bhoeas) andP.malvsefloru7n, respectively, 

 "which were growing under very adverse conditions, that 

 the polymerous androecium of the normal flower was 

 reduced to four stamens, thus reproducing in a very 

 interesting way the 4-mery of the corolla. 



Flowers of herb-Paris (P. quadrifolia) have occasion- 

 .ally been observed in which all whorls were 3-merous. 



Masters mentions a flower of Oypripedium in which, 

 while stamens a?- and a^ of the inner whorl were normal, 

 the staminode of the outer whorl was absent. This may 

 be aptly compared with the condition in Adactylvs whose 

 normal androecium is exactly the same as the abnormal 

 ■one in the Gypripedium just mentioned. Adactylus 

 belongs to the allied tribe Apostasiese. In Apostasia 

 the staminode has almost disappeared, having the form 

 of a filamentous structure adnate to the style. 



Finally, there are those cases in which all the 

 stamens are suppressed, inducing of course thereby 

 unisexuality of the flower. This is a progressive 

 change and not reversionary, for originally all flowers 

 were hermaphrodite or bisexual. 



Velenovsky observed otherwise quite healthy plants 

 of Banunculus acer in which all the flowers were two 

 or three times smaller than usual, and the petals of each 

 also smaller; the stamens were represented only by 

 insignificant papillge below the pistil. 



In some apetalous flowers of Mimulus luteus in 

 which the calyx had become petaloid, both corolla and 

 stamens were quite absent. There, are many cases 

 known where suppression of the androecium has 

 occurred without the advent, so far as could be ascer- 

 tained, of any special stimulus which might have 

 induced it, as in the Ranunculus above cited. 



VOL. II. 17 



